Search papers, labs, and topics across Lattice.
Qatar Computing Research Institute, Qatar
4
0
6
4
Arabic hateful memes pose unique challenges that current detection models are ill-equipped to handle, revealing critical gaps in our understanding of online hate.
A freely available mobile app is empowering users across nine languages to proactively spot and resist misinformation tactics through bite-sized, interactive learning.
RL-based post-training with Group Relative Policy Optimization (GRPO) can significantly boost the ability of thinking-based MLLMs to detect hateful memes by improving both classification accuracy and explanation quality.
Training VLMs on a unified, multilingual, multitask meme dataset reveals that robust meme understanding requires multimodal training and is highly sensitive to dataset-specific overfitting.
